Dynamic evolving neural-fuzzy inference system for rainfall-runoff (R-R) modelling

Dynamic Evolving Neural-Fuzzy Inference System (DENFIS) is a Takagi-Sugeno-type fuzzy inference system for online learning which can be applied for dynamic time series prediction. To the best of our knowledge, this is the first time that DENFIS has been used for rainfall-runoff (R-R) modeling. DENFIS model results were compared to the results obtained from the physically-based Storm Water Management Model (SWMM) and an Adaptive Network-based Fuzzy Inference System (ANFIS) which employs offline learning. Data from a small (5.6 km2) catchment in Singapore, comprising 11 separated storm events were analyzed. Rainfall was the only input used for the DENFIS and ANFIS models and the output was discharge at the present time. It is concluded that DENFIS results are better or at least comparable to SWMM, but similar to ANFIS. These results indicate a strong potential for DENFIS to be used in R-R modeling.

Performances of rainfall-runoff models calibrated over single and continuous storm flow events

Accurate parameter estimation is important for reliable rainfall-runoff modeling. Previous studies emphasize that a sufficient length of continuous events is required for model calibration to overcome the effect of initial conditions. This paper investigates the feasibility of calibrating rainfall-runoff models over a number of limited storm flow events. For a subcatchment having a moderate influence from initial soil moisture conditions, this study shows that rainfall-runoff models could still be calibrated reliably over a set of representative events provided that the events cover a wide range of peak flow, total runoff volume, and initial soil moisture conditions. This approach could provide an alternative calibration strategy for a small watershed that has a limited data length but consists of runoff events with a wide range of magnitudes. Compared to continuous-event calibration, event-based calibration appears to perform better in simulating the overall shape of hydrograph, peak flow and time to peak. However, continuous-event calibration was found to be more reliable in providing runoff volume, suggesting that continuous-event calibration should still be used when runoff volume is the main concern of a study.

Improved particle swarm optimization-based artificial neural network for rainfall-runoff modeling

This paper presents the application of an improved particle swarm optimization (PSO) technique for training an artificial neural network (ANN) to predict water levels for the Heshui watershed, China. Daily values of rainfall and water levels from 1988 to 2000 were first analyzed using ANNs trained with the conjugate-gradient, gradient descent and Levenberg-Marquardt neural network (LM-NN) algorithms. The best results were obtained from LM-NN and these results were then compared with those from PSO-based ANNs, including conventional PSO neural network (CPSONN) and improved PSO neural network (IPSONN) with passive congregation. The IPSONN algorithm improves PSO convergence by using the selfish herd concept in swarm behavior. Our results show that the PSO-based ANNs performed better than LM-NN. For models run using a single parameter (rainfall) as input, the root mean square error (RMSE) of the testing dataset for IPSONN was the lowest (0.152 m) compared to those for CPSONN (0.161 m) and LM-NN (0.205 m). For multi-parameter (rainfall and water level) inputs, the RMSE of the testing dataset for IPSONN was also the lowest (0.089 m) compared to those for CPSONN (0.105 m) and LM-NN (0.145 m). The results also indicate that the LM-NN model performed poorly in predicting the low and peak water levels, in comparison to the PSO-based ANNs. Moreover, the IPSONN model was superior to CPSONN in predicting extreme water levels. Lastly, IPSONN had a quicker convergence rate compared to CPSONN.

Rainfall-runoff modeling using dynamic evolving neural fuzzy inference system with online learning

This is an open access article under the CC BY-NC-ND license.Neuro-Fuzzy Systems (NFS) are computational intelligence tools that have recently been employed in hydrological modeling. In many of the common NFS the learning algorithms used are based on batch learning where all the parameters of the fuzzy system are optimized off-line. Although these models have frequently been used, there is a criticism on such learning process as the number of rules are needed to be predefined by the user. This will reduce the flexibility of the NFS architecture while dealing with different data with different level of complexity. On the other hand, online or local learning evolves through local adjustments in the model as new data is introduced in sequence. In this study, dynamic evolving neural fuzzy inference system (DENFIS) is used in which an evolving, online clustering algorithm called the Evolving Clustering Method (ECM) is implemented. ECM is an online, maximum distance-based clustering method which is able to estimate the number of clusters in a data set and find their current centers in the input space through its fast, one-pass algorithm. The 10-minutes rainfall-runoff time series from a small (23.22 km2) tropical catchment named Sungai Kayu Ara in Selangor, Malaysia, was used in this study. Out of the 40 major events, 12 were used for training and 28 for testing. Results obtained by DENFIS were then compared with the ones obtained by physically-based rainfall-runoff model HEC-HMS and a regression model ARX. It was concluded that DENFIS results were comparable to HEC-HMS and superior to ARX model. This indicates a strong potential for DENFIS to be used in rainfall-runoff modeling.

A statistical model for combustion resonance from a DI diesel engine with applications

Introduced in this paper is a Bayesian model for isolating the resonant frequency from combustion chamber resonance. The model shown in this paper focused on characterising the initial rise in the resonant frequency to investigate the rise of in-cylinder bulk temperature associated with combustion. By resolving the model parameters, it is possible to determine: the start of pre-mixed combustion, the start of diffusion combustion, the initial resonant frequency, the resonant frequency as a function of crank angle, the in-cylinder bulk temperature as a function of crank angle and the trapped mass as a function of crank angle. The Bayesian method allows for individual cycles to be examined without cycle-averaging|allowing inter-cycle variability studies. Results are shown for a turbo-charged, common-rail compression ignition engine run at 2000 rpm and full load.

Signalling corporate collapse using a dual classification scheme:Australian evidence

Regardless of the technical procedure used in signalling corporate collapse, the bottom line rests on the predictive power of the corresponding statistical model. In that regard, it is imperative to empirically test the model using a data sample of both collapsed and non-collapsed companies. A superior model is one that successfully classifies collapsed and non-collapsed companies in their respective categories with a high degree of accuracy. Empirical studies of this nature have thus far done one of two things. (1) Some have classified companies based on a specific statistical modelling process. (2) Some have classified companies based on two (sometimes – but rarely – more than two) independent statistical modelling processes for the purposes of comparing one with the other. In the latter case, the mindset of the researchers has been – invariably – to pitch one procedure against the other. This paper raises the question, why pitch one statistical process against another; why not make the two procedures work together? As such, this paper puts forward an innovative dual-classification scheme for signalling corporate collapse: dual in the sense that it relies on two statistical procedures concurrently. Using a data sample of Australian publicly listed companies, the proposed scheme is tested against the traditional approach taken thus far in the pertinent literature. The results demonstrate that the proposed dual-classification scheme signals collapse with a higher degree of accuracy.

Statistical models for cancer screening

This paper reviews the application of statistical models to planning and evaluating cancer screening programmes. Models used to analyse screening strategies can be classified as either surface models, which consider only those events which can be directly observed such as disease incidence, prevalence or mortality, or deep models, which incorporate hypotheses about the disease process that generates the observed events. This paper focuses on the latter type. These can be further classified as analytic models, which use a model of the disease to derive direct estimates of characteristics of the screening procedure and its consequent benefits, and simulation models, which use the disease model to simulate the course of the disease in a hypothetical population with and without screening and derive measures of the benefit of screening from the simulation outcomes. The main approaches to each type of model are described and an overview given of their historical development and strengths and weaknesses. A brief review of fitting and validating such models is given and finally a discussion of the current state of, and likely future trends in, cancer screening models is presented.

A manifold approach to face recognition from low quality video across illumination and pose using implicit super-resolution

We consider the problem of matching a face in a low resolution query video sequence against a set of higher quality gallery sequences. This problem is of interest in many applications, such as law enforcement. Our main contribution is an extension of the recently proposed Generic Shape-Illumination Manifold (gSIM) framework. Specifically, (i) we show how super-resolution across pose and scale can be achieved implicitly, by off-line learning of subsampling artefacts; (ii) we use this result to propose an extension to the statistical model of the gSIM by compounding it with a hierarchy of subsampling models at multiple scales; and (iii) we describe an extensive empirical evaluation of the method on over 1300 video sequences – we first measure the degradation in performance of the original gSIM algorithm as query sequence resolution is decreased and then show that the proposed extension produces an error reduction in the mean recognition rate of over 50%.